Presented atIPR’s 2^nd Post-War Reconstruction Strategies Conference

Hargeisa

July 20-25, 2000

AIDS Acquired Immunodeficiency Syndrome

HIV Human Immunodeficiency Virus

ICD International Cooperation for Development

KABP Knowledge, attitude, beliefs, practice

MCH Maternal and Child Health

MoHL Ministry of health and Labour

NGO Non-governmental Organisation

STD Sexually Transmitted Disease

UNAIDS United Nations Programme on HIV/AIDS

UNICEF United Nations Children’s Fund

WHO World Health Organisation

Somaliland seceded from the rest of Somalia in May 1991 after a civil war in which 60,000 people were killed and 500,000 fled across the borders to Ethiopia, Djibouti and beyond. The capital Hargeisa was systematically destroyed through targeted bombing by the forces of Siad Barre in 1988 and again suffered shelling during inter-clan hostilities in 1994-5.

Today there is an elected government and relative peace and security in Somaliland. Many former residents are returning to the country and the process of reconstruction and rehabilitation is underway. However, the conflict has left a legacy of destroyed buildings and infrastructure, the loss of all records, isolation due to the destruction of communications and a serious shortage of skilled personnel. Health services, which collapsed during the conflict, are functioning in many places, but there is a general lack of essential human and material resources.

While Somaliland remains in the process of post conflict reconstruction, most interventions have focussed on the immediate needs of the population. Until recently no international organisation had developed plans to address Human Immunodeficiency Virus and Acquired Immunodeficiency Syndrome (HIV/AIDS) at a local, regional or national level, although the World Health Organisation (WHO) has been supplying HIV testing kits to three blood banks since 1995.

HIV and other sexually transmitted diseases (STDs) have become a serious problem for many countries around the world, but most notably in Sub-Saharan Africa. Globally, seventy percent of people living with HIV or AIDS are from this region.(1) The prevalence of HIV in countries which have borders with Somaliland and where many Somali refugees currently live, is known to be high. (2,3,4,5) Statistical evidence on the extent and spread of HIV in Somaliland is seriously lacking,(6) but data from the blood banks, testing of suspected cases of AIDS and information from the few studies conducted in Somaliland (7,8), suggest that infection rates may be increasing.

It is important to have sufficiently accurate and complete data regarding the distribution and spread of STDs/HIV/AIDS in order to plan and implement an effective prevention and control programme (9,10,11). International Cooperation for Development (ICD) and WHO have collaborated to carry out an assessment of the prevalence of STDs/HIV. This is linked with a United Nations Children’s Fund (UNICEF) survey of knowledge, attitudes, beliefs and practice (KABP) in various populations and regions within Somaliland.

This assessment is the first step in engaging the international community to work with the MoHL to develop effective and integrated strategies for prevention and control of STDs and HIV. The results of the prevalence study are reported here.

To obtain baseline information on STDs/HIV/AIDS in Somaliland in order to develop effective and integrated strategies for prevention and control.

To obtain data on the prevalence of STDs and HIV to:

The study area covered four of the six regions of Somaliland – Awdal, Galbeed, Sahel and Sool. Study sites were chosen in Boroma, Hargeisa, Berbera and Las Anod, the major town in each region. The three target populations included in the prevalence study were antenatal clinic attenders, TB patients and one group of men offered STD services.

TB patients had to be Somalis, between the ages of 15 and 49 and have been resident in Somaliland for at least one year. The men sampled were between the ages of 15 and 55. Postnatal women and others between 15 and 49 years were included with antenatal women at sites where it was otherwise difficult to achieve the required sample size.

Sample sizes for each population at each site were predetermined on the basis of the percentage of the total population in the region, but with regard for feasibility of collecting the required number of samples in the time available. Where possible, sample sizes were based on those recommended for sentinel surveillance, with a minimum of 100 per site. (See Annex 1)

Testing took place over a four week period in October and November. For all study groups consecutive blood samples were collected until the predetermined sample size or the time limit was reached, or all suitable participants had been included.

Temporary laboratory facilities were set up at the health centres (MCH) to test antenatal patients for Syphilis. Patients were offered a blood test on arrival at the MCH Centre and given a number to be used throughout the study. After blood collection, patients proceeded to their usual antenatal check up or were interviewed for the KABP survey, where both studies were running . Patients proceeded to a physician, who took a history and examined patients for STDs. If test RPR testing was not completed at the site, patients were asked to return the following day for results and treatment of Syphilis if necessary. (See Annex 2)

Similar temporary facilities were set up to collect blood samples from male volunteers. Testing for Syphilis was carried out at the laboratory in Hargeisa Hospital. A physician returned to the testing site the following day with the results, to interview and treat study participants where necessary.

Blood samples were collected from TB patients in the TB hospitals and tested for HIV in a central laboratory. A numbering system was used throughout to ensure the testing was anonymous.

Blood samples were collected by venepuncture into numbered plain evacuated tubes (Becton Dickinson Vacutainer Systems Europe, France). To obtain sera, samples were allowed stand for a minimum of one hour or were spun in a hand centrifuge. All sera were tested for Syphilis using a Rapid Plasma Reagin test (Syphilis Reagin Card Test, Biotec Laboratories, Ipswich, UK), either at the blood collection site or later the same day. Laboratory forms with the study number and completed RPR result were passed the physician.

Sera were stored refrigerated and tested in batches for HIV antibodies within 48 hours of collection, using a rapid latex aggregation test (Capillus HIV-1/HIV-2, Galway, Ireland). After testing, sera were frozen and transported to Hargeisa.

At the end of the study period all RPR positive sera were retested using a haemagglutination assay to detect antibodies to Treponema pallidum (TPHA Kit, Lorne Laboratories Ltd, Reading, UK). Sera from subjects with symptoms suggestive of Syphilis (genital ulcers) were also tested by TPHA. All HIV positive samples were tested using a rapid immunochromatographic test (Abbott Determine HIV-1/2, Abbott Laboratories, Tokyo, Japan). Any samples found to be indeterminate (positive by one test only) were sent to the University of Nairobi for testing by ELISA.

Positive and negative control samples were run with each batch of tests. In addition a random selection of negative sera were retested for quality assurance purposes.

Study participants, except for TB patients, were identified by their study number and interviewed by a physician the same day, or asked to return the following day if the RPR results were not available. Interviews were based on the medical interview questionnaires (See Annex 3). Examinations were carried out only if the clinician judged it necessary and treatment followed WHO STD syndromic management guidelines. Patients were treated for Syphilis on the basis of the RPR result in combination with symptoms.

Both men and women were asked about previous operations and blood transfusions. Women were also asked whether they had undergone circumcision, and if so, what type (Pharaonic or Sunna) (12)

Participation in the study was voluntary and with informed consent for blood tests for Syphilis and physical examination. Those in the sample population found to be suffering from STD symptoms were treated, using WHO syndromic management guidelines, and given advice on risk reduction.

For reasons of confidentiality, HIV testing was done on anonymous unlinked samples, in large batches. As there is no treatment, counselling or support available for people with HIV/AIDS in Somaliland, and there is evidence of negative reactions to a diagnosis of HIV, informing people of their HIV status was considered inappropriate.

A total of 1,251 blood samples were collected and tested for HIV and Syphilis out of a proposed study size of 1,500. See Table 1.

In all, 1,178 samples were included in the analysis. These included 769 antenatal samples, 303 from TB patients and 106 samples from men offered STD screening services. Six samples found to be indeterminate for HIV using the rapid test kits were all found to be negative by ELISA. The results, by area, are presented in the Tables 2, 3 and 4.

Prevalence of Syphilis amongst antenatal attenders was found to be lowest in Las Anod (0.9%) and highest in Boroma (3.7%). Prevalence of HIV showed a similar pattern ranging from 0.4% to 1.6%, with an overall prevalence amongst antenatal women of 0.9%.

In total, 98.6% of women included in the study were interviewed by a physician. Of these, almost 31% presented with symptoms of STDs. Prevalence of STD symptoms was lowest in Boroma (22.6%) and highest in Hargeisa (35%).

Pharaonic circumcision was universal in Las Anod, but in Hargeisa 4% of women had had Sunna circumcision or had not undergone the procedure at all. Overall 43 (5.7%) women interviewed had had an operation or received a blood transfusion.

TB patients were only tested for HIV. The prevalence among TB patients in the three regions combined was found to be 4.6%, ranging from 1.9% in Las Anod to 7% in Hargeisa.

Amongst the men tested, prevalence of Syphilis was 1.9% and HIV 0.9%. The prevalence of STD symptoms was 12% and 5.6% of those interviewed had previously had an operation or blood transfusion.

In the original study protocol the proposed total sample size was 2,200, including 1,300 antenatal patients, 350 TB patients and 550 ‘high risk’/STD patients. This was reduced to a proposed total sample size of 1,500 after the start of sample collection based on two factors. The first factor was the unpredicted absence of a key member of the study team during the critical period of fieldwork. The second was the jailing of several women in Hargeisa involved in sex work. Sex workers were included in the original protocol as a vulnerable group, but this group was dropped from the study, as involvement may have led to identification and further stigmatisation.

In total 1,238 samples were collected. Of these 1,165 were included in the analysis. Results of antenatal samples from Berbera were excluded. The sample size (73) was smaller than the minimum stipulated in the protocol (100 samples) and there were some technical and reporting discrepancies.

Sample sizes for the different sentinel groups are recommended by WHO. These are 400 for Antenatal patients and 250-400 for TB patients per site over an eight to twelve week period. Due to the constraints of this study (mainly time) the sample sizes for each site were based on the estimated population of the region and the feasibility of collecting the samples within a four week period. Where possible the proposed sample sizes were equivalent to those recommended, but this was not achieved for any single site.

Many TB patients travel to the TB hospitals in Somaliland from regions of Ethiopia, where health facilities are not available. In order to more accurately reflect the HIV prevalence in Somaliland, only those patients resident in the country for a minimum of 12 months were eligible for inclusion in the study.

The overall prevalence of Syphilis in antenatal attenders (1.8%) and male study participants (1.9%) agrees with the findings of the study in 1997, in which a prevalence of 1.7% was found amongst antenatal patients in Hargeisa.(7)

The prevalence of HIV amongst antenatal patients was found to be 0.9%. The prevalence in Hargeisa of 0.7% is lower than the findings of the previous study (7). The prevalence of HIV amongst the men in this study (0.9%) is in broad agreement with the current HIV rates for blood donors of 1.4%.(12)

As expected, HIV rates amongst TB patients were higher than in the antenatal sample,.

Because the prevalence of both Syphilis and HIV is low, the sample sizes were not large enough for comparison between regions. Testing for significance is invalid if any of the numbers compared are below 5.

Over 98% of antenatal attenders and 84% of male study participants were interviewed by a physician.

Prevalence testing or surveillance can play an important role in characterising the magnitude and nature of the HIV epidemic. Such data are essential for advocacy as well as planning. Good prevalence data can be used to estimate the current magnitude of the HIV/AIDS problem and assist decision-makers in implementing appropriate and rational intervention programmes.(8)

HIV surveillance can be defined as the collection of epidemiological data on the distribution and spread of the virus, which is sufficiently accurate and complete to be relevant to the planning and implementation of interventions to control the spread of HIV/AIDS. Continuing surveillance monitors trends in HIV infection and allows projections to be made.

Sentinel surveillance is the method of choice for obtaining data on HIV infection rates in various population groups and monitoring trends. Sentinel surveillance alone, however, does not provide enough information to explain changing patterns of infection or to evaluate effectiveness of interventions. Therefore trends in infection need to be monitored alongside trends in behaviours which can lead to infection.(11)

It is also recognised that the presence of untreated STDs increases the risk of acquiring HIV up to tenfold.(12) As there is little accurate information available on the prevalence of STDs in Somaliland, the study design includes testing for Syphilis and syndromic management of other STDs.

1. UNAIDS. AIDS Update. Geneva December 1998
2. UNAIDS/WHO. Epidemiological Fact Sheet on HIV/AIDS and sexually transmitted diseases: Somalia. June 1998
3. UNAIDS/WHO. Epidemiological Fact Sheet on HIV/AIDS and sexually transmitted diseases: Djibouti. June 1998
4. UNAIDS/WHO. Epidemiological Fact Sheet on HIV/AIDS and sexually transmitted diseases: Ethiopia. June 1998
5. UNAIDS/WHO. Epidemiological Fact Sheet on HIV/AIDS and sexually transmitted diseases: Kenya. June 1998
6. Rodier, G et al. Infection by the human immunodeficiency virus in the Republic of Djibouti: literature revue and regional data. Med Trop (Mars)1993;53(1):61-7
7. Ahmed, HJ. STD/HIV prevalence and chemotherapy studies in Somaliland and Somalia. Report 1998 (WHO/SIDA)
8. Salama, P. Report on a mission to North West Somalia (Somaliland) and the Somali Region of Ethiopia (Region 5): situational analysis of STDs/HIV/AIDS. September 1997 (UNAIDS/UNHCR)
9. EMRO. A Practical Guide to HIV/AIDS Surveillance. WHO-EM/STD/11/E/G
10. UNAIDS. Guidelines for HIV interventions in emergency settings. UNAIDS/96.01
11. Mayaud, P. et al. STD rapid assessment in Rwandan refugee camps in Tanzania. G U Med 1997;73(1):33-38
12. Duffy, G. Report of STD/HIV/AIDS Assessment for the Ministry of Health and Labour, Somaliland. November 1999.
13. UNAIDS. Guide to the strategic planning process for a national response to HIV/AIDS: Situation Analysis. UNAIDS/99.21E
14. EMRO. Guidelines for Conducting an Assessment of the Prevalence of Sexually Transmitted Diseases in women Attending Antenatal and Gynaecology Clinics. WHO-EM/STD/15/E/G 1998